JP4663590B2 - Peak position variation measuring apparatus, measuring method and program thereof - Google Patents

Description

  The present invention relates to a peak position variation measuring apparatus, a measuring method thereof, and a program, and more specifically, a peak position variation measuring apparatus used for biomolecular interaction measurement used in data analysis and biochemical research by spectroscopic analysis. , Its measuring method and program.

  Spectrum, the wavelength dispersion SPR spectra, incident angle dispersion SPR spectra, QCM (Quartz Crystal microbalance), cantilever Laver mass meter, the peak or valley such chromatogram (hereinafter, referred to as the peak) in the spectrum having a peak position of the two spectra And the fluctuation of the peak position of the spectrum is measured.

  Substances are identified using the wavelength of a peak in spectroscopic analysis and the peak time in a chromatograph. In the SPR measurement using the surface plasmon resonance effect, the refractive index is obtained from the dip position seen in the reflection intensity spectrum with respect to the wavelength of the absorption spectrum or the incident angle due to the SPR effect.

  In general, in a chemical analysis method using a resonance effect that causes a peak in a spectrum, it is necessary to measure the conditions under which resonance occurs. In particular, in SPR, the resonance condition is obtained from the reflection intensity with respect to the incident angle and the reflection intensity with respect to the incident wavelength, and the refractive index is obtained from the incident angle and the incident wavelength of the resonance condition. In an apparatus using an array-type photosensor as a light receiving element, the reflection intensity is obtained as a discrete spectrum with respect to the incident angle or incident wavelength, and the obtained spectrum is processed by a computer to obtain a final index of refraction and intermolecular interaction. Is derived.

  In such a technique, the measurement resolution greatly depends on the processing method. In particular, the SPR method has a single peak, and by obtaining the position (referred to as the D axis) representing the energy or momentum, the refractive index change can be obtained with high precision. Therefore, it is necessary to determine the amount of movement about the D axis from data including drift.

  As a method for obtaining the movement amount related to the D axis, for example, the shape of the peak is approximated by a polynomial, the polynomial is determined by a regression method (FIG. 4A), or the graphical center of gravity of the shape of the peak is obtained ( FIG. 4B) (see Patent Document 1 and Non-Patent Document 1) and a method of determining a peak position with high accuracy by a weighted regression method (see Non-Patent Document 2).

Japanese Patent No. 3490458 Knut Johansen, Ralph Stalberg, Ingemar Lundstrom and Bo Liedberg, “Surface plasmon resonance: instrumental resolution using photo diode arrays”, Meas. Sci. Technol., Vol. 11, IOP Publishing Ltd, 2000, pp.1630-1638 Kyle S. Johnston and Sinclair S. Yee, “Calibration of Surface Plasmon Resonance Refractometers Using Locally Weighted Parametric Regression”, Analytical Chemistry, Vol. 69, No. 10, May 15, 1997, pp. 1844-1851 Peter A. Gorry, “General Least-Squares Smoothing and Differentiation by the Convolution (Savitzky-Golay) Method”, Analytical Chemistry, Vol. 62, 1990, pp.570-573

  However, the conventional method alone has a problem that the accuracy of determining the peak position is limited by noise due to non-uniformity of the light source, distortion of the optical device, and uneven sensitivity of a light receiving device such as a CCD or CMOS camera.

  In addition, in the processing method using a computer, the input spectrum has a discrete number of sample points, and the peak displacement is less than or equal to the sample point interval particularly in high-sensitivity measurement using SPR. For this reason, in the measurement of a minute change in refractive index and the concentration of a substance having an extremely low concentration, there is a problem that the change in the peak position is buried in noise and cannot be measured.

  Furthermore, when determining a representative position narrower than the discrete interval for a discrete spectrum, the conventional one-step method, such as performing polynomial fitting to determine the value of the extreme independent variable, selects the data to be used for fitting. In some cases, the representative position obtained in the above changes discontinuously, causing problems in measurement.

  The present invention has been made in view of such a problem, and an object of the present invention is to provide a peak that can measure a change in the representative position of a peak with high accuracy at an interval narrower than the discrete interval with respect to a discrete spectrum. The object is to provide a position variation measuring device, a measuring method thereof, and a program.

In order to achieve such an object, the invention according to claim 1 is one-dimensional pixel column data representing a spectrum obtained by measuring a first sample at a predetermined time interval in a predetermined optical system. Spectrum generating means for generating certain first discrete series data in time series, and second discrete data that is one-dimensional pixel string data representing a spectrum obtained by measuring a second sample in the optical system. First storage means for recording series data; parallel movement means for reading the second discrete series data from the first storage means and translating it in a column index direction by a predetermined parallel movement amount; a representative position calculation means for calculating a representative position of the peak for each of the discrete-series data and second discrete series data moved parallel, and parallel movement amount of the second discrete-series data and the representative position An array generation means for generating an array of pairs, a second storage means for recording the array, an array is read from the second storage means, and a relational expression between the translation amount and the representative position is obtained, and the relation Conversion means for converting the representative position of the first discrete series data into a corresponding translation amount at an interval narrower than the discrete interval of the first discrete series data from the equation, and translation of the first discrete series data It is characterized by comprising difference calculation means for arranging the amounts in time series and taking and outputting the difference of the parallel movement amount in a specific period.

  A second aspect of the present invention is the peak position variation measuring device according to the first aspect, wherein the representative position calculating means converts the first and second discrete series data to either Fourier transform or wavelet transform. And a function representing the first and second discrete series data is obtained analytically by calculating the phase of the frequency component having the longest period.

  A third aspect of the present invention is the peak position variation measuring device according to the first aspect, wherein the spectrum represents a reflection intensity with respect to an incident angle for each pixel arranged in one dimension.

  The invention according to claim 4 is the peak position variation measuring device according to claim 3, wherein the spectrum is a spectrum in which a peak of light absorption due to the surface plasmon resonance effect appears.

The invention according to claim 5 is a peak position variation measuring method, which is one-dimensional pixel string data representing a spectrum obtained by measuring a first sample at a predetermined time interval in a predetermined optical system. A step of generating a certain first discrete series data in time series, and a second discrete series data which is one-dimensional pixel sequence data representing a spectrum obtained by measuring a second sample in the optical system In the first storage means, the second discrete series data is read from the first storage means and translated in a column index direction by a predetermined translation amount, determining a representative position of the peak for each of the second discrete-series data that has been moved discrete series data and parallel, and parallel movement amount of the second discrete-series data and the representative position A step of generating an array for generating an array of pairs, a step of storing the array in the second recording means, and reading out the array from the second storage means to obtain a relational expression between the translation amount and the representative position; A step of converting a representative position of the first discrete series data into a corresponding parallel movement amount at an interval narrower than a discrete interval of the first discrete series data from the relational expression, and a parallel of the first discrete series data A step of arranging the movement amounts in time series, and taking and outputting a difference between the parallel movement amounts in a specific period.

  The invention according to claim 6 is the peak position variation measuring method according to claim 5, wherein the step of calculating the representative position includes Fourier transform and wavelet transform to the first and second discrete series data. This is a step of analytically obtaining a function representing the first and second discrete series data by performing any of the above and calculating the phase of the frequency component having the longest period.

The invention according to claim 7 is a program for measuring a peak position variation, which is one-dimensional pixel string data representing a spectrum obtained by measuring a first sample at a predetermined time interval in a predetermined optical system. A step of generating a certain first discrete series data in time series, and a second discrete series data which is one-dimensional pixel sequence data representing a spectrum obtained by measuring a second sample in the optical system In the first storage means, the second discrete series data is read from the first storage means and translated in a column index direction by a predetermined translation amount, representative determining a representative position of the peak for each of the parallel movement amount of the second discrete-series data of the second discrete-series data that has been moved discrete series data and parallel A step of generating an array for generating an array consisting of a pair with a position, a step of storing the array in the second recording means, and a relational expression between the parallel movement amount and the representative position by reading the array from the second storage means And converting the representative position of the first discrete series data into a corresponding translation amount at an interval narrower than the discrete interval of the first discrete series data from the relational expression, and the first discrete series The step of arranging the parallel movement amounts of the data in time series and taking the difference of the parallel movement amounts in a specific period and outputting them is performed.

  According to the present invention, even in a minute change in refractive index that could not be measured because the change in peak position was buried in noise, the measurement of the concentration of an extremely low concentration substance, etc., the discrete spectrum is narrower than the discrete interval. It is possible to measure the variation of the peak representative position with high accuracy at intervals.

  When the amount of peak movement is small, such as fluctuations in the peak position that are smaller than the sample interval of the spectrum, in general, the spectrum often moves parallel to the D axis without changing the peak shape. Therefore, in the present invention, when obtaining the fluctuation of the peak position smaller than the sample interval of the spectrum, the spectrum obtained as one reference spectrum is translated, and the peak representative position obtained from the spectrum and the series of translation amounts Is used to find the fluctuation of the peak position from the representative position of the inspection spectrum obtained in time series. That is, according to the present invention, the representative position is determined from the relationship between the representative position obtained by translating the spectrum obtained as the reference spectrum and the translation amount, in contrast to the conventional method for obtaining the representative position of the spectrum in one step. There is a feature in the point to ask.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
(Embodiment 1)
FIG. 1 shows a configuration of a peak position variation measuring device according to an embodiment of the present invention. The peak position variation measuring device 100 includes a spectrum input device 101 and a processing device 102 connected by a serial interface.

  The spectrum input device 101 measures the SPR curve (spectrum) of the sample by Handy-SPR (NTT-AT). Specifically, the spectrum input device 101 measures the reflected light intensity with respect to the incident angle with a one-dimensional CCD, AD converts the light intensity of each pixel of the one-dimensional CCD, and forms an SPR curve as a string of 12-bit integer values. Output the data. That is, the data of the SPR curve measured by the spectrum input device 101 is given as a discrete expression. The SPR curve data is output from the spectrum input device 101 to the processing device 102 via a serial interface.

  The processing device 102 derives the fluctuation of the peak position from the SPR curve data output from the spectrum input device 101. In this embodiment, the index (pixel number) of the spectrum column is the axis D. The processing device 102 processes the reference spectrum S1 and the inspection spectrum S2 output from the spectrum input device 101 in order to improve measurement accuracy. As the reference spectrum S1, it is desirable to use a spectrum in which the difference between the representative position of the peak of the reference spectrum S1 and the representative position of the peak of the inspection spectrum S2 is within ± 1 pixel. In addition, the reference spectrum S1 is not necessarily obtained by measuring the exact same sample, but the reference spectrum S1 may be obtained by measuring a sample immediately before measurement to obtain the inspection spectrum S2. desirable. That is, even if the reference spectrum S1 and the inspection spectrum S2 are generated separately or obtained by measuring collected samples, they are measured by the same optical system, and the peak representative positions of those spectra are shifted by one. As long as it is about.

  The representative position calculation units 103A and 103B for obtaining the representative position of the peak perform Fourier transform or wavelet transform on the pixel string data to represent the spectrum as a periodic function and calculate the phase of the frequency component whose number of samples is the period. To calculate the representative position of the peak. The representative position calculation units 103A and 103B have exactly the same function, and both the reference spectrum S1 and the inspection spectrum S2 may be processed by one representative position calculation unit 103A or 103B.

  The translation unit 104 that translates the pixel column data moves the axis D by a positive or negative translation amount, for example, repeatedly executes the process of moving the pixel column data to the adjacent pixel by one pixel. The reference spectrum S1 ′ moved for each process is generated. In this way, by moving the spectrum sequence data in parallel, noise fixed to the wavelength superimposed on the spectrum due to imperfectness of the device that measures the spectrum (such noise is usually considered to be a device function). Can be canceled. This is particularly effective when determining a representative position narrower than the discrete interval for a discrete spectrum.

  Further, the translation unit 104 outputs the generated reference spectrum S1 ′ to the representative position calculation unit 103A, and outputs the translation amount to the array generation unit 105.

  The array generation unit 105 creates a pair of the representative position of the reference spectrum S1 ′ obtained by the representative position calculation unit 103A and the parallel movement amount, creates an array A, and stores it in the storage device 108.

  The conversion unit 106 for obtaining the corresponding parallel movement amount from the representative position takes out the array A of the reference spectrum S1 corresponding to the inspection spectrum S2 from the storage device 108, and represents the peak of the inspection spectrum S2 output from the representative position calculation unit 103B. A translation amount corresponding to the position is derived. When obtaining the translation amount corresponding to the peak representative position of the inspection spectrum S2, the conversion unit 106 may use a relational expression between the peak representative position and the translation amount derived from the array A using the least square method. it can. By this method, the shift amount of the peak of the discrete spectrum can be determined narrower than the discrete interval regardless of the property of the representative position calculation unit 103.

  The difference calculation unit 107 calculates a difference from the reference, with the parallel movement amount of each inspection spectrum S2 output from the conversion unit 106 as a predetermined reference parallel movement amount, for example, and calculates the difference in time series. Can be output side by side to a PC. In addition, for example, a Savitzky Golay method in which moving average or differentiation of series data is performed by weighted addition of data, such as performing noise removal on these time series data, can be used (Non-patent Document 3).

  Such a processing device 102 includes a CPU that executes control, a ROM that stores a control program for the CPU, and a storage device 108 such as a RAM that provides a work area for the CPU and stores data. This can be realized by adopting a configuration controlled by a control unit (not shown) provided. However, the present invention is not limited to this, and any technique can be used as long as it is a technique used in a normal personal computer. Further, the processing apparatus 102 includes an input operation unit (not shown) including a keyboard or various switches that are connected to the control unit and inputs predetermined commands, data, and the like, and stores a large amount of data such as image data. An auxiliary storage device (not shown) can also be provided.

  FIG. 2 shows a measurement method of the peak position variation measuring device according to one embodiment of the present invention. FIG. 3 shows a measurement example by the peak position variation measuring device according to one embodiment of the present invention. Data of the reference spectrum S1 is input from the spectrum input device 101 to the translation unit 104 (S201). However, since the reference spectrum S1 may be measured in advance, it may be input to the translation unit 104 from a predetermined storage device (not shown) instead of being output from the spectrum input device 101. The translation unit 104 generates a reference spectrum S1 ′ in which the data of the reference spectrum S1 is moved to an adjacent pixel by an arbitrary pixel (S202), and obtains a representative position of the peak of the translated reference spectrum S1 ′. The data is output to the representative position calculation unit 103 A and the parallel movement amount is output to the array generation unit 105. The representative position calculation unit 103A calculates the representative position of the peak of the translated reference spectrum S1 ′ (S203). Next, it is determined whether or not the translation amount of the reference spectrum S1 'is within a preset range (S204). If the translation amount is within the range, the process returns to S202, and if it is outside the range, the reference spectrum S1 is determined. The representative position of the peak of ′ is output to the array generation unit 105. The array generation unit 105 generates an array A including a pair of the representative position of the peak of the reference spectrum S1 ′ and the corresponding parallel movement amount, and stores the array A in the storage device 108 (S205).

  On the other hand, the spectrum input apparatus 101 measures the inspection spectrum for measuring the fluctuation of the peak position (S206), and inputs the inspection spectrum S2 from the spectrum input apparatus 101 to the representative position calculation unit 103B for obtaining the representative position of the peak (S207). The representative position calculation unit 103B calculates the representative position of the peak of the inspection spectrum S2 (S208), and outputs it to the conversion unit 106 that calculates the corresponding parallel movement amount from the representative value. The conversion unit 106 obtains a relational expression between the peak representative position and the translation amount from the array A stored in the storage device 108 by the least square method or the like, and obtains the translation amount corresponding to the peak representative position of the inspection spectrum S2. Calculate (S209).

  It is possible to obtain a time series of the peak positions of the inspection spectrum S2 by sequentially performing operations on the inspection spectrum S2 obtained in time series by repeating S206 to S209. Then, the difference of the parallel movement amount of each inspection spectrum S2 is taken with reference to the parallel movement amount of the predetermined inspection spectrum S2, and the difference is arranged in time series and output to a PC or the like (S210). In S210, noise can be removed using a Savitzky Golay method or the like in which moving average or differentiation of series data is performed by weighted addition of data.

  Since the reference spectrum S1 is taken from the same device as the inspection spectrum S2, noise and fluctuations originating from the device are not woven into the array, so that noise originating from a measuring device such as an optical system can be canceled it can. Furthermore, random noise that cannot be canceled by this mechanism can be removed by performing general processing such as the Savitzky Golay method in the difference calculation unit 107. Therefore, in the embodiment of the present invention, the temporal variation of the peak position of the inspection spectrum can be measured more precisely.

  The apparatus of the present invention can also be realized by a computer and a program, and a program for functioning as a peak position variation measuring device can be recorded on a computer-readable recording medium or provided through a network. .

It is a figure which shows the structure of the peak position variation | change_quantity measuring apparatus which concerns on one Embodiment of this invention. It is a figure which shows the measuring method of the peak position variation | change_quantity measuring apparatus which concerns on one Embodiment of this invention. It is a figure which shows the example of a measurement by the peak position variation | change_quantity measuring apparatus which concerns on one Embodiment of this invention. (A) is a figure explaining calculating | requiring a representative position by approximating a peak shape with a polynomial, and (b) is a figure explaining calculating | requiring a representative position from the graphical center of gravity of a peak shape.

Explanation of symbols

  100 Peak position fluctuation measuring device

Claims (7)

Spectrum generating means for generating, in time series, first discrete series data that is one-dimensional pixel string data representing a spectrum obtained by measuring a first sample at a predetermined time interval in a predetermined optical system;
First storage means for recording second discrete series data, which is one-dimensional pixel string data representing a spectrum obtained by measuring a second sample in the optical system;
Translation means for reading the second discrete series data from the first storage means and translating it in a column index direction by a predetermined translation amount;
Representative position calculating means for obtaining a peak representative position for each of the first discrete series data and the translated second discrete series data;
Array generating means for generating an array composed of pairs of parallel movement amounts and representative positions of the second discrete series data;
Second storage means for recording the arrangement;
The second from the storage unit reads the sequence sought relationship between the representative position and amount of translation, the first discrete closely spaced than the discrete distance between the first discrete-series data from the relational expression Conversion means for converting a representative position of the target series data into a corresponding parallel movement amount;
A peak position variation amount, comprising: difference calculating means for arranging the parallel movement amounts of the first discrete series data in time series and taking and outputting the difference of the parallel movement amounts in a specific period. measuring device.   The representative position calculating means performs any one of Fourier transform and wavelet transform on the first and second discrete series data, and calculates the phase of the frequency component which is the longest period, thereby calculating the first and second 2. The peak position variation measuring device according to claim 1, wherein a function representing the discrete series data is analytically obtained.   The peak position variation measuring device according to claim 1, wherein the spectrum represents a reflection intensity with respect to an incident angle for each pixel arranged in one dimension.   The peak position variation measuring device according to claim 3, wherein the spectrum is a spectrum in which a peak of light absorption due to a surface plasmon resonance effect appears. Generating, in time series, first discrete series data that is one-dimensional pixel array data representing a spectrum obtained by measuring a first sample at a predetermined time interval in a predetermined optical system;
Recording in the first storage means second discrete series data that is one-dimensional pixel string data representing a spectrum obtained by measuring a second sample in the optical system;
Reading the second discrete series data from the first storage means and translating it in a column index direction by a predetermined translation amount; and
Obtaining a peak representative position for each of the first discrete series data and the translated second discrete series data;
Generating an array that generates an array of pairs of parallel movement amounts and representative positions of the second discrete series data;
Storing the arrangement in a second recording means;
The second from the storage unit reads the sequence sought relationship between the representative position and amount of translation, the first discrete closely spaced than the discrete distance between the first discrete-series data from the relational expression Converting the representative position of the target series data into a corresponding translation amount;
Arranging the parallel movement amounts of the first discrete series data in time series, and taking and outputting a difference between the parallel movement amounts in a specific period.   The step of calculating the representative position performs either the Fourier transform or the wavelet transform on the first and second discrete series data, and calculates the phase of the frequency component that is the longest period, thereby calculating the first and second discrete series data. 6. The peak position variation measuring method according to claim 5, wherein the peak position variation measuring method is a step of analytically obtaining a function representing the second discrete series data. Generating, in time series, first discrete series data that is one-dimensional pixel array data representing a spectrum obtained by measuring a first sample at a predetermined time interval in a predetermined optical system;
Recording in the first storage means second discrete series data that is one-dimensional pixel string data representing a spectrum obtained by measuring a second sample in the optical system;
Reading the second discrete series data from the first storage means and translating it in a column index direction by a predetermined translation amount; and
Obtaining a peak representative position for each of the first discrete series data and the translated second discrete series data;
Generating an array that generates an array of pairs of parallel movement amounts and representative positions of the second discrete series data;
Storing the arrangement in a second recording means;
The second from the storage unit reads the sequence sought relationship between the representative position and amount of translation, the first discrete closely spaced than the discrete distance between the first discrete-series data from the relational expression Converting the representative position of the target series data into a corresponding translation amount;
A program for measuring a peak position fluctuation amount, comprising: arranging parallel movement amounts of the first discrete series data in time series, and taking and outputting a difference between the parallel movement amounts in a specific period. .

Images